Factors Affecting the Career Choices
of Medical Students and Residents
Presented by :
Martey S. Dodoo PhD.
Robert Graham Center,
Washington, DC
Academy Health 2009 Annual Research Meeting
June 28 – 30, 2009
Chicago, IL
Background




Policy: Physician training  fewer primary care
compared to sub-specialists
Students: Declining popularity of primary care
Payment models  imbalance – PC vrs non-PC
Imbalance  increased costs & underserved areas
So now more than ever, we need to understand:
1.
Why students not choosing primary care
or underserved practice
2.
Where physician workforce can best be utilized
to benefit greater society
Previous Research
Effect of:
 Student factors
 Debt (mixed)
 Loan repayment programs
 Institutional factors (3)
Knowledge gaps & research question

Major knowledge gaps in previous research:
 Debt & Interaction
 Scholarship & Loan repayment
 Student’s relative expected income
Research Question:
 Do any of the following help predict specialty choice or practice
location?
 Level of student debt?
 Exposure to Title VII funding?
 Student’s relative expected income?
 Returns to investment in medical education?
 Exposure to primary care or underserved training
Data sources







Graduation survey (GQ) (1979–2004) (AAMC)
AMA Master file (2001 –05)
Medicare outpatient claims -health centers (2001-05)
NHSC participant database
Title VII exposure data base (1979–2004)
Primary care med. residency history data (1970-2004)
Supplementary data
Methods




Quality checks of all data files
Match/merged data files  1 analysis file
For each set of variables:
 SAS (v9.2) calculations
 Logistic regressions
Estimated average hours-adjusted NPV
(ROI in medical education)
Data merge/match diagram
AAMC GQ
AMA
Master File
Survey Debt Data
(1998-2004)
Physician Data
(2001-2006)
310,705
710,988
Medicare claims in
FQHC, RHC etc. (2001-2005)
174,310 visits
68,106 physicians
155,221 beneficiaries
NHSC History & Title VII
funding
- NHSC service
- NHSC scholarship
obligation
- Med School Title
Funds
Excluded
- Ostepathic physicians
- International Grads (IMG)
- Graduated Med school before
1979 or after 2001
- Graduated residency before
1982 or after 2004
388,857
Analysis Data File
317,260
Residency
History & Title VII
funding Data
VII
542,939
Geographic
Data
- RUCA
codes
Basic Findings
Physician Distribution in analysis file (n=317,260)
NHSC obligation & service
Practice location
NHSC scholarship
1.7% Practice in MUP
3.4%
NHSC loan payment oblig.
0.7% Practice in MUA
14.6%
Ever NHSC service
2.3% Practice in Popln. HPSA
11.1%
Practice in Geog. HPSA
3.1%
Career choice
Family Physicians
13.2% Rural practice
9.8%
General Internists
13.3% FQHC practice
4.7%
Gen. Pediatricians
7.6% RHC practice
5.8%
Not primary care
65.9%
Ever Primary care
35.3%
More findings

Title VII exposure associated with:
FM & und. electives  FM & underserved practice






Inner City, Rural, Primary Care Clerkships & Electives Matter
Rural birth 
2.4 x rural practice
1.8 x Family medicine practice
Public Med School  1.8 x FM & Rural practice
Men half as likely to  primary care practice
Women half as likely to  rural practice
Married docs 50% more likely to  Family Med. practice

Effect of debt:



More findings
With no debt less likely to  rural, primary care practice
Up to $150K debt more likely  rural, PC practice
Over $150K debt  Likelihood of practice (above) declines

Doctors trading debt for service  up to 7 X more likely to
choose primary care & underserved area practice

FQHC or RHC practice:


Enhanced by obligating scholarship.

Reduced by having debt.
Title VII exposure associated with increased rural practice

Enhanced by an obligating scholarship.
More Findings
Trends in payments to physician groups(1979-2003)
$450,000
Diagnostic Radiology
$400,000
$300,000
$250,000
$200,000
$150,000
$100,000
Primary Care
$50,000
20
03
20
01
19
99
19
97
19
95
19
93
19
91
19
89
19
87
19
85
19
83
19
81
$0
19
79
Median Annual Income
$350,000
More Findings
Average hours-adjusted Net Present Value (NPV) (5%)
and Prim. Care vrs. non-PC Difference in Lifetime earnings:
Primary
care NPV
Not prim.
Care NPV
Difference in total
lifetime earnings
1980-1989
$12.54
$36.58
$1,751,724
1990-1999
$23.43
$59.95
$2,845,379
2000-2004
$29.58
$78.45
$4,093,931
Sources: Analyses from the Robert Graham Center, using data from AAMC,
JAMA, Bureau of Labor Statistics, MGMA
Study Limitations



We could not obtain data on:
 Family income,
 Parent professional status,
 Career interests
We did not find data on Osteopaths
Pediatricians bill Medicare infrequently (only
disabled beneficiaries).
Policy Implications

To promote primary care & redistribute physician workforce to
underserved areas may have to:

Reauthorize and revitalize Title VII

Shift training  underserved area communities.

Increase primary care payments (cut disparity w/ non_PC payments)


Admit students more likely to choose primary care or practice in
underserved areas.
Provide more “debt for service” programs
What’s needed:
Future studies on what makes underserved areas attractive practice
locations, especially for women
Our Study Team
Acknowledgements
This study was supported by The Josiah Macy Jr. Foundation
The authors also wish to thank:
Association of American Medical Colleges (AAMC)
Diane Rittenhouse (UCSF)
Scott Shipman (Dartmouth)
Ed Fryer (University of Arkansas)
Sarah Brotherton (AMA)
Candice Chen (GWU)
Fitzhugh Mullan (GWU)
David Goodman (Dartmouth)
Contact email:
mdodoo@aafp.org